Good Weighing Practices rn5h

Accurate & Efficient Weighing Good Weighing Practice in Modern Industry Erik Volek – Lab Sales Consultant in Upstate NY and Western PA Jack Blakley – Industrial Business

Development Manager

Accurate and Efficient Weighing Greeting and Introduction Quick history of weighing / preparing the weighing side of our minds

Weighing Matters

5’

5’

Accurate weighing measurements matter in every industry

Life Cycle Management of Weighing Equipment All segments must be considered from design qualification to routine performance testing

Accuracy, Measurement of Uncertainty and Minimum Weight

5’

15’

Review of Accuracy, MoU, and Minimum Weight

Minimum Weight and Safety Factor – Establishing Accuracy

10’

Establishing a low end accuracy limitation

Performance Verification – Assuring Accuracy How and how often should a scale be tested? Can costs be reduced without compromising compliance?

10’

The Revolution

1946 The first METTLER scale that was developed following the substitution principle:

Only one weighing pan instead of two! Thank you Dr. Erhard Mettler

Evolution: The E M F R Weighing Cell

1973: The first time presented to the public: The electronic precision balance METTLER PT1200. Electro Magnetic Forced Restoration

Electromagnetic compensating analytical balance 1 weighing pan 2 hanger 3 guide 4 flexible t 5 coupling 6 lever 7 lever bearing 8 compensation coil 9 permanent magnet 10 magnetic flux 11 optical position sensor 12 position vane 13 temperature sensor

5

Copyright 2009 Mettler-Toledo, Inc.

Accurate & Efficient Weighing Weighing Matters

Dave Cirullo GWP Consultant

Accurate weighing: Essential for your processes

Guaranteed Accuracy – Reduced Costs – Secure Audits  We want to produce high-grade products.

 We want to save costs and reduce expenses.  We want to meet ISO and other regulations.

How can your weighing systems help and meet the above requirements? Avoid the following pain points…. 8

Avoid These Pain Points

Prevent bad batches and OOS*?

Problem source: Inaccurate weighing -QUALITY-

Prove scales are Missing adequate consistently accurate to documentation the auditor! -COMPLIANCE-

Appropriate testing at appropriate intervals

Testing is costly and a lot of effort -COST-

Good Manufacturing Practice “Say what you do, do what you say, and be able to prove it.”

Courtesy Novartis

Good people

Courtesy Roche

+

Good equipment

+

10

Courtesy Nocartis

Good processes

=

Courtesy Nocartis

Good products

FDA Warning: Risk management

Risk management

Weighing Device is Calibrated; is it appropriate? I have a Calibration Report from my service provider, but are my balances and scales appropriate for my weighing process?

"Mr./Mrs. Customer: I have assessed, recorded data, and performed adjustments on all of your weighing devices. They are within manufacture's specifications. Here are the calibration reports for your file." -Qualified Service Provider

"My SOP says I need to weigh 1kg of this key ingredient"

Capacity: 3'000kg Readability: 1kg

± 80% uncertainty

0.4 kg

1.5kg

What you see is not what you get! Each measurement comes with measurement uncertainty. In order to weigh in a reproducible manner, this should be known.

How do you guarantee Accurate weighing? Wrong ideas:  “What you see is what you get”  Inaccurate measurement results  "I test my scales every day"… but incorrect or redundant tests cause unnecessary costs

Accurate & Efficient Weighing Life Cycle Management

GWP® - The Global Weighing Standard

Metrology



Quality management  Process requirements 

Life Cycle Management The weighing community should have a science-based standard for efficient life-cycle management of weighing equipment. Clear answers should be available on: how you calibrate each device, how you establish of limits, and sound scientific rationale covering every step in the life cycle management of a weighing device

New Equipment Evaluation

The requirements for the scale are documented transparently and scientifically. In particular the most important are: - The smallest and largest initial weight - Accuracy requirements

- Quality standards to be met - Don't buy a device based readability alone

New Equipment Selection

Which scale can weigh 200 g with an accuracy of at least 1 %?

200g/0.01mg

4000g/0.01g

60kg/1g

600kg/50g

The knowledge about the measurement uncertainty and the minimum weight is the requirement for the selection of a suitable, accurate measurement instrument.

Installation by a trained technician

A professional installation and qualification ensures good measurement results from the beginning.

Training assists s in the correct handling of your measuring instruments.

Calibration of each weighing device

The determination of the measurement uncertainty and the minimum weight ensures that the instrument supplies measurements with the required accuracy.

The weighing device is calibrated and appropriate for the weighing process

Routine Operation / Performance Checks  Appropriate testing during the entire life cycle of the weighing system assures accurate weighing measurements

 Establish scientific rationale for routine testing, based on the risk of the weighing process:

-

How is the weighing instrument to be checked? How often? To what extent can the expenses for tests be reduced?

 Minimization of risk and cost savings are achieved by testing what is required. This way good weighing results are consistently achieved.

The Measurement Uncertainty of a Weighing Device

Measurement Uncertainty – what is it? “Measurement uncertainty […] Parameter that describes the dispersion of the quantity values that is assigned to a scale […].” International Vocabulary Of Basic And General In Metrology (VIM) JCGM 200:2008, entry 2.26

 The measurement uncertainty takes of of the circumstance that no measurement is perfect

 The measurement uncertainty indicates how accurate the measured value is – how close it is to the destination

 The indication of the measurement uncertainty is a part of quality systems, such as ISO, GLP/GMP or pharmacopoeia such as the USP

 The accuracy of measurements can be checked during an inspection and needs to be documented

Example: Quality Assurance in the formulation We would like to formulate 100 kg of fragrance in a drum with a weight of 20 kg. We need the following ingredients:

 Water :  Oil:  Emulsifier:  Moisturizer:  Stabilizer:  Fragrance:

70 % 20 % 5% 3% 1,5 % 0,5 %

 500 g

Each component is to be weighed with an accuracy of 1 %.

If possible, we would like to work with a single scale. Which scale do you recommend?

Example: Quality Assurance in the formulation

Can I use a scale with a readability of 100 g to weigh a substance having a weight 500 g with an accuracy of 1 %?

 500 g ± 5 g

This scale can indicate 500 g; is it still accurate to ± 5 g? Of course not!

Accuracy = Readability? Or do I need a scale with a higher readability, such as 10 g or even 1 g?

But what about the accuracy?

Accuracy  Readability

Display: The readability of a scale can be electronically set to any value.

Sensor: To increase the accuracy, a better sensor must be selected (e.g. EMFR instead Strain Gauge technology)

The accuracy of a scale is expressed by its measurement uncertainty.

What Accuracy Means

A scale is ACCURATE if it satisfies the requirements of the . In other words: MEASUREMENT UNCERTAINTY < PROCESS TOLERANCE

Accurate, reliable Measurement results Measurement uncertainty of m2

Measurement uncertainty of m3

Process tolerance

Measurement uncertainty of m1

OK

Fair m1 kg

Not OK m2 kg

m3 kg

Display [kg]

Accuracy means that the MEASUREMENT UNCERTAINTY of the instrument is always less than/equal to the PROCESS TOLERANCE.

Measurement Uncertainty – what is it? “Measurement uncertainty […] Parameter that describes the dispersion of the quantity values that is assigned to a scale […].” International Vocabulary Of Basic And General In Metrology (VIM) JCGM 200:2008, entry 2.26

What are the main contributors to uncertainty of measurement in a weighing device?

Measurement Uncertainty of a scale nominal property:

Readability

RD

Measurement properties:

Sensitivity

SE

Non-linearity

NL

Corner load

EC

Repeatability

RP

All these properties contribute to the measurement uncertainty of the scale.

The perfect Scale

Measurement uncertainty By readability, …

(RD)

Sensitivity Deviation, …

(SE)

Non-linearity, …

(NL)

Corner Load/Eccentricity/Shift… (EC)

Repeatability

(RP)

Determination of The measurement uncertainty Calibration of non-automatic electronic measurement instruments:

U

 2 

v

w



v

r





v W e

2





v W a

Measurement uncertainty

U

Weight

W

Deviation of the repeatability

Vw

Rounding error deviation

Vr

Deviation of the corner load

Ve

Deviation of the non-linearity

Va

Effect of the deviation of the

VK

Calibration weight

2



v

K



W

2

Measurement uncertainty [kg or %]

Measurement Uncertainty behavior

Weight [kg]

Max.

Relative measurement uncertainty [%] = Absolute measurement uncertainty / weight

Focus: Minimum Weight and Safety Factor Establishing Accuracy Limitations

Measurement uncertainty %]

Accuracy limit = Minimum weight

Process tolerance [%] (Accuracy requirements of the customer)

Weight [kg]

Max.

Accuracy limit: Minimum weight

If there is weighing below the minimum weight, the measurement uncertainty is greater than the required accuracy:  Inaccurate results

Measurement uncertainty %]

Accuracy limit = Minimum weight

Process tolerance [%] (Accuracy requirements of the customer)

Weight [kg]

Max.

Accuracy limit: Minimum weight

When weighing above the minimum weight, the measurement uncertainty is always smaller than the accuracy required: Accurate results

Measurement uncertainty %]

Minimum weight increases as process % gets tighter

Minimum weight for 1%

1%

0.5% 0.1% Minimum weight for 0.5%

Minimum weight for 0.1%

Weight [kg]

The greater the accuracy, the tighter the process tolerance the higher the minimum weight

The Concept of Minimum Weight in USP GC 1251  "The minimum net sample weight, mmin, of a balance can be expressed by the equation: mmin = k*s/required weighing tolerance

 For materials that must be accurately weighed, Balances 41 stipulates that repeatability is satisfactory if two times the standard deviation of the weighed value, divided by the net desired smallest weight, does not exceed 0.10%. For this criterion the equation above simplifies to: mmin = 2000*s

 If not subject to the requirements of Balances 41, the minimum weight value may vary depending on the required weighing tolerance and the specific use of the balance."

This methodology describes an alternative for determining the minimum weight of balances. 45

Internal usage only

Test Weight for Repeatability/Minimum Weight? Indication

At the lower end of the measurement range, the repeatability is almost constant.

Smaller weight: Feasible, but difficult to handle (especially for micro and analytical balances)

Test weight up to a few percent of the balance's capacity.

Load « 5%

5%

Internal usage only

The Concept of Minimum Weight in GC1251  "In order to satisfy the required weighing tolerance, when samples are weighed the amount of sample mass (i.e., the net weight) must be equal to or larger than the minimum weight. The minimum weight applies to the sample weight, not to the tare or gross weight."

The net sample weight has to fulfill the minimum weight requirement.

GC 1251 wants to overcome the misconception that a tare weight can be used to fulfill the minimum weight requirement. 47

Internal usage only

Minimum Weight can change…use a Safety Factor  "Factors that can influence repeatability while the balance is in use include:

- The performance of the balance and thus the minimum weight can vary over -

time because of changing environmental conditions Different operators may weigh differently on the balance—i.e., the minimum weight determined by different operators may be different The standard deviation of a finite number of replicate weighings is only an estimation of the true standard deviation, which is unknown The determination of the minimum weight with a test weight may not be completely representative for the weighing application. The tare vessel also may influence minimum weight because of the interaction of the environment with the surface of the tare vessel."

 "For these reasons, when possible, weighings should be made at larger values than the minimum weight."

The above information stipulates the application of a "safety factor". 48

Internal usage only

What is the Safety factor?

Safety factor < 1 Weighing results are inaccurate and process fails

Safety factory Between 1 and 2 Weighing process is temporarily functioning properly, but the reproducibility cannot be guaranteed over the long term. Safety factor ≥ 2 (depends on risk)

Accuracy ensured over the long-term

Variability of the Minimum weight – safety factor Minimum weight Smallest net weight Adjustment (by Service)

SAFETY AND SECURITY

Calibration range

Calibration range

Time Initial calibration

"As Found" calibration

"As Left“ calibration

Traceable Documentation of Minimum Weight

What is the minimum weight for an accuracy of 1%?

Minimum Weight Certificate

Safety Factor: Assures Accuracy over time

We recommend using a safety factor >1 to assure accuracy over time. Minimum Weight Certificate

Focus: Performance Verification Ensuring accuracy

Test strategy for the Measuring equipment monitoring Calibration: Evaluation of the overall performance of the scale by testing all relevant weighing parameters, including integrated safety functions (e.g. automatic calibration)

 Instrument Routine testing: Evaluation of the requirements for the weighing accuracy

 Weighing process FACT: Compensation of external (temperature) and internal effects (drafts)

 Environment Risks during weighing are reduced and accurate results at minimal costs are ensured by the combination of the 3 test procedures.

Calibration by Service technician

Benefits  Evaluation of the overall performance of the scale by testing all relevant weighing parameters  Determination of the measurement uncertainty and the minimum weight  Output of recognized and verifiable certificates according to ISO 17025/USP/A2LA Restrictions

 Unknown status of the scale between several calibrations if they were not frequently executed

 High costs if they were frequently executed

Self-testing by Intelligent scales FACT Without

With

Benefits

 Automatic adjustment compensates for drafts and corrects temperature effects

 Reduction of costs and effort, because on scales having FACT fewer routine tests are required. Restrictions

 Weights are formally not verifiable for national or international standards and guidelines.

 No determination of repeatability or corner load errors

Functionality of Auto-Calibration / Adjustment  At the end of the assembly process in the factory, the builtin weight of the balance is being calibrated with an external reference weight (“primary reference calibration”).

 The value of this calibration is stored in the memory of the balance. The built-in weight is used to calibrate or adjust the sensitivity of the balance during any subsequent auto-calibration / adjustment.

57

Copyright 2009 Mettler-Toledo, Inc.

Functionality of Auto-Calibration / Adjustment  Internal adjustment mechanisms are a great compliment to a weighing quality process

 They should not be used in place of a routine testing

 They should be used in conjunction with routine testing

 They should be challenged by appropriate routine testing performed by end s and appropriate service providers

 For most weighing equipment,

58

only OEM service technicians have the ability to adjust these mechanisms Copyright 2009 Mettler-Toledo, Inc.

Routine testing by

Benefits

 Effortless and rapid checking of the required accuracy

 Immediate determination of deviations  Minimization of risks by testing at appropriate intervals between the calibrations

Restrictions

 Incomplete testing of the properties of the scale

 Weights must be recalibrated at regular intervals

GWP: A risk-based Test strategy

Weighing accuracy

Impact

Risk Management Which checks should I perform to effectively test my measuring instruments and minimize risks?

Let us first consider the properties that actually affect your weighing results, i.e., that mainly contribute to measurement uncertainty. We recommend that you check only these relevant properties.

Measurement Uncertainty of a scale Nominal property:

Readability

RD

Measurement properties:

Sensitivity

SE

Non-linearity

NL

Eccentricity load

EC

Repeatability

RP

All these properties contribute to the measurement uncertainty of the scale.

Suggested Uncertainty Testing

Relatively expanded measurement uncertainty [%]

Effect of the weighing Parameters on the accuracy Scale XP 204: Individual effects on the uncertainty (@k = 2)

Sample mass [g]

Repeatability dominates uncertainty

Sensitivity and corner load dominate uncertainty

Summary: Influence of Weighing properties

Repeatability Dominant influence on samples with low mass

Sensitivity and eccentricity load Dominant influence on samples with high mass; Influence on samples with low mass negligible

Non-linearity Not dominant with samples having high mass; generally not dominant over the entire sample mass range

Normally it is not required that s test the non-linearity. This test is performed in the context of the service technician.

Which weight for The sensitivity test?  Test the sensitivity with a weight that corresponds to the capacity (weight 1).  The use of a low test weight leads to a lower detectable sensitivity deviation. The measurement is influenced in part by repeatability (pink marking).

 If an even smaller test weight is used, the sensitivity deviation is completely covered by the repeatability. Characteristic curve with Sensitivity deviation

Display

Correct sensitivity

Weight 1: Recommended

Incorrect Weight 2: Permissible

Load « 5%

5%

Nominal capacity

Weight 2 for Testing the repeatability Display

At the bottom end of the measuring range, the repeatability is nearly constant.

Lower weight: Permissible, but difficult in handling

Weight 2: Recommended

With load « 5%

5%

The assessment of the repeatability is much easier with weight 2 than with a much lower weight (weight 2 is less prone to operating errors).

Test weights For tests Weight 1: Maximum OIML/ASTM weight, that is less than or equal to the capacity of the scale (except scales with high capacity) Weight 2: Maximum OIML/ASTM weight that is less than or equal to 5 % of the capacity of the scale

XP205 200 g and 10 g

PBA430x-60 50 kg and 2 kg

Accurate & Efficient weighing processes Questions to ask a quality manager…

4 Simple Questions to Answer Are my balances/scales accurate?

Is our documentation complete?

Am I performing too many or too few tests?

Do I have an overview of the status of all devices?

Answers you should hope to receive….  Analysis of the scales and their accuracy in the process

 Recommendations for measuring equipment monitoring based on risk analysis

 Complete documentation  Overview of the status of all scales

 Training

GWP® - The Global Weighing Standard

Metrology



Quality management 

Process requirements 

IIndustrial

Application questions Jack Blakley, [email protected] (800) 638-8537 x7200